PRECISION SPINDLE METROLOGY Pt. 2: Sensitive Direction

[@mikemac5070]

Omg, babe, wake up. New precision metrology video just dropped

[@danf6070]

Hey, hoping adding my comment will help with the algorithm bringing this video to more people's feeds. I've been following your channel for a long time and I thoroughly enjoy your content. I'm a self-taught machinist who produces parts for scientific equipment, but I'm intensely fascinated by the theories behind why the things we observe during machining are the way they are. You're one of the only channels on here that is getting this granular with the background science and I find it absolutely riveting. Thanks for all your hard work and keep it up! This content is incredibly valuable.

[@cylosgarage]

Thank you for the kind words

[@electrowizard2000]

Mastefully explained. It's one of those things I never would have guessed is true, but once it's properly explained is totally obvious. Thanks!

[@cylosgarage]

Thanks! I’m glad it made sense

[@adamthemachinist]

excellent video , never thought about the outside diameter to bore roundness before

[@davidbraden7689]

Great video, with really nice demonstrations and models.

[@charliemcmonagle]

These are great. I am going to wait and watch this properly with a notepad. I work in scientific research at a central facility with loads of metrology but not from a formal engineering background. This is a great resource and your videos have been very helpful and currently I am looking to have an air spindle installed on the instrament so thank you again.

[@stevenhavener7327]

I am certainly Not measuring My spindle, that would be a rabbit hole I can not afford to go down !! Best regards, Steve ( love the content BTW )

[@jimsvideos7201]

8:58 That is indeed pretty bananas.

[@GermanMythbuster]

I love learning such cool metrology stuff, even though I will probably never need it. But it help you thinks about engineering problems differently. So glad you made it approachable for smooth brain apes like me 😅 Love you videos so much

[@cylosgarage]

Thanks very much. Don’t assume im not a smooth brained ape too amigo

[@MF175mp]

Monkey likes shiny

[@paulclose]

Mind Blown!!!

[@evzone84]

As a crude hobiest, this is way out side what im ever considering. But an amazing explanation for something that's not very intuitive. Nice work.

[@raindeergames6104]

Lovely stuff😊

[@creepysneeze1]

This is good shit bro

[@lumotroph]

Clever🎉

[@Aleksandar_Sladic]

Awesome!

[@YulehHabibi]

hello cyrus. Nice video blog.

[@janbeck8269]

This blew my mind! Thanks for sharing! Does professional instruments mark their spindle where this optimum happens? I do wonder how impactful this can be in reality when non synchronous error motion gets added. But you kind of said you'd address this in the next episode...

[@cylosgarage]

They can determine and provide that information if requested but don’t do it by default. Due to grejda optimization mostly being used for air bearing spindle applications, the asynchronous error doesn’t end up affecting it very much because the asynchronous error motion is so tiny for air bearings.

[@joshclark44]

It's weird how close the algorithm comes to mind reading sometimes. I was just thinking about this the other day after watching a video of someone making a collet chuck for his spindle (probably what actually informed the algorithm) and he was talking about runout with 3 jaw, 4 jaw, and collet chucks and how they change depending on how you mount them. Even with a collet chuck, you have to sometimes (in his case) tap them into place to get near perfect runout, and I was wondering how cool it'd be if you could permanently add some sensor to the spindle to measure the vibrations, per theta, in x and y, and have some display for indicating the balance. You should be able to quickly tap them into place, but also if you have a part on a faceplate that isn't perfectly balanced, you could tell in which direction to add additional weight to balance it, and how much. You could tap the counterweights for the part into the right spot to make it perfectly balanced and have it run smoothly every time (assuming nothing slips and the center of mass as you remove material remains constant etc)

[@raindeergames6104]

In your humble opinion do You think it is worth while to take take a relatively small home lathe (weight is about 250kg chinese machine) and upgrade the spindle bearings to a better class? My plan is to do a full cnc convertion on this machine with Centroid controller.

[@cylosgarage]

It depends on what you want to do with it. Practically, a “normal” lathe is likely not going to be limited by its spindle error motion. Unless it’s truly terrible. Where we draw the line is when the error motion is in the same order of magnitude of the tolerances we’d like to achieve. Ideally it’s kept to less than 10 percent of our tolerance. Unless you are trying to turn to sub-tenth tolerances, or using an MCD tool or something, the gains from better bearings may be negligible. But if you’re really unhappy with its current performance, give it a go

[@graham854]

This old dining room table.

[@un2mensch]

I was like "oh shiiiiiii... oohhhh shiiiii...!! my guy aboutta flip a couple trig equations for some swept manifolds right on up to outer space into the complex plane! I'm gonna blow!" then he go and hide that shit behind a big ol' MATLAB

[@Leo.Wolf.the.Engineer]

Great video! I love the Matlab tool. Can you share the code? I'd love to play with that :P

[@cylosgarage]

@@Leo.Wolf.the.Engineer yea shoot me an email and I’ll send it over. Thanks !

[@Trainwreck1123]

I don't understand one comment you made near the end of the video. How can the error motion for the entire circle be calculated from just measurements at 0 and 90 degrees?

[@koharaisevo3666]

The error motion cannot be calculated it has to be measured over the entire rotation what can be calculated is the error motion from a particular fixed sensitive direction by measure the error motion from 2 fixed sensitive direction in x and y axis by the fomular at 10:25. 11:37 you measure the wiggly line on the left put your indicator 90 degrees, measure the wiggly line again then you can calculate all other wiggly line from other direction.

[@EverettWilson]

Practically, any motion at any degree will have some degree of X and some degree of Y. An error at one degree will have a little bit of X and a lot of Y. Same concept with two degrees, three, four, etc. Math wise, X and Y are independent and can describe any vector from 0 degrees to 360.

[@Trainwreck1123]

​@@EverettWilson Am I understanding correctly that we're talking about continuous measurements at two points, and not two instantaneous (I.E. spindle not moving) measurements? I think that might have been the cause of my confusion.

[@EverettWilson]

@@Trainwreck1123 Ah! Yes.

[@henery2000]

Can you post the matlab code?

[@lohikarhu734]

Maybe I'm incorrect, but the error motions may not be truly synchronous if there is any play, or flex in the system, as harmonics will affect the amount and direction of some, maybe most, of the motion, and these errors will be sensitive to speed, and mass "concentricity". Strictly "errors", in terms of runout of bearings or concentricity of a shaft, are only accurate and repeatable at "DC", that is, at rotational speeds low enough that flexture in the system is not "excited"? Also, of course, the radius of the part has a direct effect on the amount of tangential error at the 90° point of measurement, I think? I'm just a real old test and measurement guy.

[@ivprojects8143]

First! Excited to watch.

[@realcygnus]

nifty

[@lucaswilkins9217]

Yeah, turned and bored errors being different is definitely not obvious.

[@kuchervano]

It should be illegal for you to make us wait till next weekend for part 3😮

[@JaenEngineering]

I dunno. Having everything on demand and being able to binge kinda kills the fun. I think the anticipation adds to the excitement as the time for the next episode approaches and lead to a much more enjoyable experience. Plus it kinda makes me feel a little nostalgic for the old days of TV.

[@Hydrogenium1]

Interresting video, but maybe check your thumbnails. directly below this video i get a recommendation, coming from you, having "pt1" in its description and "pt2" in its thumbnail. that being said, kudos for hard metrology for the people.

[@cylosgarage]

Bozo strikes again. Thanks for the heads up

[@rafusy]

Awesome!